My Experiences with Music Composition

Having had several discussions about music composition with family members, I would like to talk about my experiences with trying to compose music, and how I evolved over 30 years. In the previous blog entries you can find links to some Youtube recordings for reference. Incidentally, I've completed part 3 of the Haunted Dreams series I've written, last song is titled Thelxiepeia's Lament. I started writing this 1/19/2013 and completed this on 3/23/2013.

Context: I took 10 years of piano and solfege lessons, but never took a class in music composition. I decided this was a journey I had to take on my own.

Snippet from someone's email conversation with my dad: I told my teacher I didn't want to just learn whatever was the current fad (as there are many fads even in composing).  I wanted to learn the whole art over many generations.  His belief was, to be a fluent composer you needed to know and write in all periods so that you can use any element in any style and that the future of music will include the entire repertoire of expression (from other cultures as well as the western tradition).  I agreed wholeheartedly.

My response: I agree in principal but this is hard to accomplish. I don't think I'll ever be able to write some of the stuff that Mozart wrote. I admit the larger your database of patterns is, the more flexibility you have. That's why I've definitely developed "my thing" over 30 years, but not that much beyond that unless I actually tried to sit down and play it so that I would at least get a feel for the style.

The same way a multi-prong approach to practicing piano helps, a similar multi-prong approach to music composition works as well. For example, in piano practice, helpful techniques include continuity rule (breaking things down bar by bar plus the next note and stringing them together), practicing the beginning and end of the piece first and early, starting in the middle and continuing, waking up in the morning starting cold and playing from start to finish without stopping, learning to play through your mistakes, etc. Practice them all together and you vastly increase your chances of being able to successfully play through a live performance, no matter what happens.

So how did I apply a multi-prong approach to practicing music composition? Go back to the 3/2012 blog entry about Peisinoe. What I saw in the TED talk and started to apply myself was one technique for practicing music composition. Study different styles, randomly pick notes and try to compose in that style.

  • Try varying time signatures.
  • Try chords, sequences of notes.
  • Try picking just white notes randomly.
  • Later (maybe much later) add the whole chromatic scale and try to pick random notes and compose from that. I found that to be much richer and much harder to do.

If you do that frequently, say once a day, you'll quickly build music composition skills.

Before I saw the TED talk I started with singing to the radio. Start with singing the melody, then harmonizing in simple thirds. Unfortunately it's a large leap from this to the next which is: try harmonizing something completely different without ever singing the same note as the melody. Do that on a regular basis and that will improve your ability to improvise on the fly. Last: try harmonizing to music you've never heard. I can do it because when you do it long enough you realize it's like crossing a river. If you try to cross the entire river you won't get there, but if you realize there are 5 stepping stones in front of you and you can hop on 2, you'll see you then have three choices. Make a choice and from there you can see the next 5 stones so you hop, make a random choice, and from there you can see the next 5, and eventually you get to the other side.

If you're not sure where this unfamiliar piece of music is going to go next, add a rest instead of a note for the first beat. As soon as you hear the first note or chord, you immediately know where the music is heading (your 5 stones and choices are visible again) and you just keep hopping along.

And even if you guess wrong, it going off kilter is perfectly fine if it resolves later in that bar, or further down the line.

More deliberate composing sessions:

  • When you first start out you're not very good and don't have good control. Just explore different chords, sequences, chord progressions.
  • You usually have no coherence, you'll jump from one thing to another and what melodies you do create will initially be short and wander.

Your goals should be to begin building your own database of patterns, of chords and progressions you like that make sense, things that you can begin to string together. As time passes, work to make those patterns longer and less repetitive.

I've learned never to end a day of composing by sticking a temporary bandaid on a composition. I've ruined some pieces that way. Leave it as ambiguous as possible, maybe it's an up or down sequence that can lead to many different options. Then sleep on it and it will usually resolve into something nice in a day or so.

Snippet from email:

We would also differ on the point: "My present view about the theory of music is that it belongs in the realm of neuroscience, and that every conservatory needs a department of neuroscience of music". Yes perhaps you are right about the  current way music theory is taught but, generally that way of teaching misses the boat completely.  Many times Mr. Sheinfeld used to say "Music, unlike children, is made to be heard, not seen."   I cannot see how the study of neuroscience in a conservatory is going to yield much. What would yield great benefits is to compare the ways great works of art, visual, literary, musical, architecture, etc. are structured.  Herman Hesse wrote a book called "The Glass Bead Game" based on the idea that great works of any idiom share a mastery of unity, contrast and design which are universal.  And I can still only get glimmers of that kind of understanding.

My response: I don't know enough about this topic, except to say if the questions are biological in nature, we must surely at this point know what centers of the brain light up when we experience something pleasurable vs. not pleasurable, so just from a standpoint of running our brains through tests while listening to music should yield results. Are they absolute, I'm curious. What if some kid is used to hearing horrible stuff, so when he grows up it lights up the pleasure centers of his brain because it reminds him of growing up? I don't know.

Snippet from email:

My feeling (and this may be wrong -- please tell me) is that today's "music theory" is not theory at all but is what scientists call a "reverse-engineering" using existing music to find ad-hoc rules that they seem to follow.

Yes. Quite a lot of it is that -- or even "deconstruction."  However when Beethoven undertook to study strict traditional counterpoint as it was taught in his time he submitted to all these rules with the idea that, once he understood them, he would be entitled to revise them -- and he did do that knowingly.  

My response: True. I'm afraid that's all we can do is to keep analyzing all these different patterns. But the giants all started with nothing and then created something that hadn't existed before. The only thing I can say is that is what drove me batty for 30 years straight. And it's not like I'd hear these wonderful things but boo hoo I don't have perfect pitch so I can't write them down. (I knew from age 7 when I started piano and solfege that I had perfect pitch.) Since I was a child I felt something was missing in the music world, and all those years were experimentation followed by process of elimination to cull out the undesirable traits until finally after cutting all the cruft away I figured out what I was trying to achieve. The end result was Peisinoe, part 1 of the Haunted Dreams series. (Incidentally, at age 48 I continue to grow - part 3 smokes part 1, and I can't imagine how these skills will continue to grow and mature).

As I said to my sister, in my case "my thing" involves searching for a deep richness that isn't jarring, by mixing major (left hand)/minor (right hand) or minor (left hand)/augmented (right hand) chords and progressions.

I find it interesting also that when I started and didn't have good control, I would compose to a point and get a little nervous thinking: OK, we've done our thing, time to change gears (whether a key change, time signature change, new melody, switching from a series of notes to chord progressions, etc). So initially those changes came from fear, but after awhile I began to realize it had the potential to turn into my greatest strength, and I think it has.

Another interesting thing to practice because it has happened for both Sue and me - music progresses from start to finish and most people assume that's how it's composed but many times you want to go back to the beginning or to some expansion of what was played at the beginning, so learning to compose something that joins up with some future melody is a good thing to practice. I certainly haven't mastered that, and that's why it took me four years to finish Peripathetique. I kept wandering until it finally resolved itself. Sue's Tarantella was like that, I remember she wrote the ending long before she completed the song because the perfect ending just popped into her head.

Fusion is another thing. Punk Bach or Punk Polka. Beethoven done in Ragtime. The possibilities are endless, and what new things might come from that are endless too.

Bottom line: I don't think you can just decide you want to become a composer and become a good composer. I suspect the good composers are driven in the way that I was, that they see or hear something that they don't see manifested in the world and they are compelled to do whatever they have to, for however long, to manifest that vision into reality. So the only thing I think you can talk about are the techniques that will help facilitate that process.

Exchange through email:

Is there any commonality between what I wrote and what music theory treats today?
I really don't know what music theory treats today, at least at universities and conservatories.  But I have heard from current composition students that music schools have their favored way of composing (style) which becomes so dominant that other ways of exploring sound/music are discouraged. And that reflects the way theory is presented. My own teacher developed his ideas directly from the scores of the masters and had an encyclopedic memory of those scores.  We would sit down to talk of a musical idea and he would reach up for a score of one of the masters and pinpoint a phrase from memory, and then go and get another two or three examples from other scores, within moments finding the relevant quotes.  And yet to listen to the music he composed you would think more of the sounds of a Boulez or Eliot Carter.  Not all he wrote touched me, but what he taught me from the past allowed me to take my own path -- not follow his or any others.

My response: I would love to know what she learned. I agree that it would be much too easy for me to know "my thing" and force it onto others. I don't have a broad enough background to know how to teach principles without forcing my vision onto others. Well, not true. Then you do what the Philip Glasses of this world do. Play with every conceivable musician you can find under the sun. Each one opens your eyes to a whole new world of notes, rhythms, culture, etc. And it eventually infuses into your database of patterns.

So how do you balance finding "your thing" while also studying other forms of expression?

There are two approaches:

  1. You're convinced you see something others don't so you work toward finding that, even though it may take awhile. After you've figured that out, then you expand and add to your core essential style.
  2. You explore all different styles first, and in the process figure out what you like/don't like/want to do.

The other thing is you don't want to squash or kill anything in a student, especially if it doesn't fit into norms and could end up being something different, just because no one sees it currently. Unfortunately it's difficult to distinguish between something that sounds bad because a person has an intent but doesn't have the skill to manifest that intent properly yet. Versus someone who plain isn't musical.

So there's a bit of a delicate chicken and egg problem. I chose the longer route because I was stubbornly insistent that what I was looking for, no one could teach me. I had to find it myself first and understand it before I was ready to expand to evaluating other styles as a composer myself.

At least three more subjects:

Writer's block. When you first start off there may be a lot of things you start but don't finish, I even came up with a couple things that started in the middle, no beginning or ending.
Practicing endings is a good thing, that took me a long time to figure out.

1. When you do get writer's block, it helps to learn to sleep on it.

  • When that doesn't work, it's possible it's not just a lack of inspiration, you might be blocking yourself. Why?
  • It's getting too repetitive, you want something different but not sure how. Different key? Different rhythms? Going from series of notes to chords, vice versa?
  • Sometimes you might have to back up and look for previous branches you missed, one will allow you to continue the flow whereas another branch will be stop and go and frustrating.

The other main reason I get writer's block: I can come up with stuff but it's trite, and I know I can do better, a lot better. I'll hold off until that better thing comes along.

Probably the best way to counter: practice going from anywhere to anywhere, in any shape or form. A high school classmate opened my eyes to this when I wrote something on the fly and as usual couldn't come up with a good ending. With his guitar and composition training, he promptly burped out 4 perfectly valid endings, none of which I could follow on the fly.

Different instruments inspire in different ways. Fragments played in different keys yield completely different compositions. Changing a key temporarily might yield different insights.

2. Learn to be your own guinea pig

There are a whole other level of skills you need to learn when you become a music composer because unlike established works with hundreds of years of learning do's and don'ts, you as the composer have none of that benefit.

  • Don't ever settle for something assuming you'll improve on it later. You probably won't and you will have practiced into a deep groove that will make unlearning difficult. Minimize the amount of unlearning you will have to do in terms of what you play, fingering, etc. Follow the same rules of consistency with fingering, know where you came from, where you're going, and make sure it's all cohesive and works. Then and only then should you practice what you've written, shooting for performance level skill.
  • Velocity. When you're composing something and can't play it, sometimes it's hard to know if it's correct. If you have something like MuseScore and are writing the score as you compose, you'll be able to hear it at full speed and can add/subtract notes if you need to get from point A to point B faster or slower. Do you need to change the time signature? Repeat something? Remove a repeated section?
  • Usually I focus on composing and don't worry about writing the score simultaneously. That's good and bad. It's good because you leave yourself unfettered (hmm, maybe I don't want to do that because it's yet Another key change/time signature change I'm going to have to grapple with and my dictation stinks). Tough. Write it without regard to how hard it will be to score it later.

Part 2 of the Haunted Dreams series was an exception. I decided since I didn't write anything down for 30 years I should try and write this one as I composed it. The result of that was sometimes I would write it in MuseScore before I could play it, and I would discover it was a lot harder to play than I'd anticipated. That's another thing that happens as you get better and better at composing - you'll reach a point where you compose things you can't play.

There's probably a lot more on both subjects but it's a start.

The other annoying thing that happened to me for years is I would experiment, write something, and it would be the coolest thing I'd written so far, and I would make another evolutionary leap and look back and not be able to continue the old compositions. I would leave them and start fresh. Also, in the early days I found I was profoundly affected with disconnects: I would compose and keep composing until I was tired and hungry and cranky. Next day with fresh eyes and a full belly and feeling content, I would find it impossible to continue where I left off in such a different emotional mood.

When a former piano student couldn't believe I could just write new age stuff on the fly and I said it was easy, I made a point and cranked out 16 little pieces or snippets, I did them simultaneously. As I came up with an idea I would go as far as I could, record it, move on. I would get stuck there, go back, or start a new one. That's another good system to practice.

Also practice going back later when you're good, even for much earlier things, and finishing them. I did that with compositions I started in 1982.

3. Recording can be very annoying initially when you're learning to compose.

You'll write something awesome and totally forget it when you try to repeat it. That gets better after years of practice. Also, you're afraid initially that you'll write the perfect thing and forget it (I've done that many times). Later you don't care. If it's really that good, it will surface again later, or if not quite there, it will resurface in a far better and more mature form. If it's meant to be or not meant to be, don't fight it.

But initially I tried to record everything. Hours of useless pattering around. What you really need is a looping recorder where you can set a time interval to your style, say you want to be able to capture a minute's worth of composing. So you patter around uselessly until suddenly it clicks and something comes out. Then you click the button and the recorder captures the last minute of recording so you can write down the music and continue. That would be helpful for beginner composers so they're not as afraid of losing stuff.

Major relief. I can now check this off my to do list. :)